Each year, 33-43% of breast cancers (>100,000 cases) exhibit de novo resistance to initial therapeutic treatment strategies. Unfortunately, current technologies that guide initial treatment choices for breast cancer patients are inaccurate for identifying de novo resistance. This underscores the critical clinical need for earl, accurate, and cost-effective methods of selecting the most effective initial treatment for breast cancer patients. Early identification of those tumors that will respond to therapy versus those that are resistant will (1) expedite clinical decisions regarding the course of treatment, (2) improve the clinical outcomes of breast cancer patients by identifying those patients who are in need of alternate therapies, and (3) spare pre-identified unresponsive patients from the toxicities associated with ineffective treatment. The central goal of this proposal is to develop an innovative platform based on optical metabolic imaging technologies to directly measure de novo resistance of primary tumors and predict therapy response. Optical metabolic imaging (OMI) includes a unique combination of variables, developed by our lab, to accurately measure early drug response with cellular resolution. OMI exploits the intrinsic fluorescence intensities and lifetimes of the metabolic co-enzymes reduced nicotinamide adenine dinucleotide (NADH) and flavin adenine dinucleotide (FAD) to probe cellular metabolism. Our innovative combination endpoint, the OMI index, and unique cellular-level analysis provide unprecedented sensitivity to identify heterogeneous cellular drug response, which is critical to ensure that no malignant cells escape treatment. We have shown that OMI can accurately measure therapeutic response in breast tumors in vivo at an earlier time-point than currently used clinical techniques. Further, we have developed novel methods to culture primary human breast tumors in a three-dimensional collagen matrix (organoids), and have applied OMI to accurately, rapidly, and reproducibly predict in vivo tumor responses to multiple treatment schemes in these organoids. The novel and powerful predictive value of organoid testing and imaging could be used to optimize clinical treatment strategies prior to treatment using ex vivo biopsy samples. Therefore, the significance of OMI of primary tumor organoids lies in its potential to measure the dynamic cellular response to multiple treatment strategies within individual patient tumors. This approach could predict an optimal, individualized treatment strategy within hours, before the patient is actually treated. Significance: OMI of organoids derived from primary human tumors could serve as an accurate predictive test of de novo drug resistance across multiple types of cancers, thus transforming patient care to identify optimal treatment strategies before treatment is initiated. This work also has the potential to significantly accelerate pre-clinical drug discovery by developing sensitive in vivo measures of treatment response, and a high-throughput platform to test tumor response to multiple treatment schemes while reducing animal burden and read-out time.